Pathogenic, versatile and tunable activity of sortase, a transpeptidation machine
Wójcik, Magdalena
DOI:
10.33612/diss.119637108
IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite from
it. Please check the document version below.
Document Version
Publisher's PDF, also known as Version of record
Publication date:
2020
Link to publication in University of Groningen/UMCG research database
Citation for published version (APA):
Wójcik, M. (2020). Pathogenic, versatile and tunable activity of sortase, a transpeptidation machine.
https://doi.org/10.33612/diss.119637108
Copyright
Other than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).
Take-down policy
If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.
Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons the number of authors shown on this cover page is limited to 10 maximum.
Magdalena Wójcik Susana Váquez Torres Wim J. Quax Ykelien L. Boersma
University of Groningen, Groningen Research Institute of Pharmacy, Department of Chemical and Pharmaceutical Biology, A. Deusinglaan 1, 9713 AV Groningen, The Netherlands.
Under revision in Protein Engineering, Design & Selection
SORTASE MUTANTS WITH
IMPROVED PROTEIN
THERMOSTABILITY AND
ENZYMATIC ACTIVITY
OBTAINED BY CONSENSUS
DESIGN
5
MagdaBinnenwerk.indd 77 MagdaBinnenwerk.indd 77 19/02/2020 13:53:5919/02/2020 13:53:5978
ABSTRACT
Staphylococcus aureus sortase A (SaSrtA) is an enzyme that anchors proteins to
the cell surface of Gram-positive bacteria. During the transpeptidation reaction performed by SaSrtA, proteins containing an N-terminal glycine can be covalently linked to another protein with a C-terminal LPXTG motif (X being any amino acid). Since the sortase reaction can be performed in vitro as well, it has found many applications in biotechnology. Although sortase-mediated ligation has many advantages, SaSrtA is limited by its low enzymatic activity and dependence on Ca2+. In our study, we evaluated the thermodynamic stability of the SaSrtA wild type and found the enzyme to be stable. We applied consensus analysis to further improve the enzyme’s stability, while at the same time enhancing the enzyme’s activity. As a result, we found thermodynamically improved, more active and Ca2+-independent mutants. We envision that these new variants can be applied in conjugation reactions in low Ca2+ environments.
Keywords
Staphylococcus aureus Sortase A, Conjugation, Protein engineering, Consensus
design, Protein stability.
MagdaBinnenwerk.indd 78
79
5
INTRODUCTION
The bacterial transpeptidase sortase A (SrtA) enzyme is found in the membrane of most Gram-positive bacteria, where it mediates the anchoring of proteins and virulence factors to the cell wall. The best characterized and most studied SrtA is that of Staphylococcus aureus (SaSrtA).[1,2] This enzyme recognizes a C-terminal sortase-specifi c motif, which consists of the fi ve amino acids LPXTG (X being any amino acid), and cleaves the peptide bond between the penultimate (Thr) and last (Gly) amino acid. After cleavage, the substrate remains bound to the enzyme until this intermediate is resolved by a nucleophilic attack from a second substrate, containing an N-terminal pentaglycine motif.[3,4] Interestingly, the sortase-mediated reaction can be used in vitro for site-specifi c conjugation reactions,[5,6] however, the soluble version of SaSrtAΔN59 wild type (WT) lacking the fi rst 59 residues is a catalytically ineffi cient enzyme.[7–10]
Both directed evolution and rational design have been employed to optimize certain enzyme characteristics.[11–14] The choice of strategy depends on the availability of a robust high-throughput screening or selection method, and of data on the enzyme’s structure-function relationship; naturally, both methods can be used simultaneously in a semi-rational approach.[15] So far, SaSrtA
ΔN59 has been successfully engineered to create mutants with an altered substrate specifi city[16] as well as catalytically more active mutants using directed evolution.[9,10,17] An evolved SaSrtA
ΔN59 pentamutant (PM) was reported to have a 120-fold higher kcat/Km LPETG and a 20-fold higher Km GGG.[9] However, a higher enzymatic activity sometimes comes at the expense of stability,[18] which can have implications for reuse, immobilization or storage.[19] The stability of SaSrtAΔN59 WT was previously characterized using circular dichroism.[20,21] We also used far-ultraviolet circular dichroism (CD)[22,23] and determined the reversibility of thermal unfolding of both truncated SaSrtAΔN59 WT and SaSrtAΔN59 PM subjected to increasing temperatures. We found that both enzymes are thermodynamically stable proteins, with a reversible change in CD as a function of temperature. Previous eff orts to improve the melting temperature (Tm) include linking two intradomain cysteine residues using biselectrophiles.[24] Here, we applied consensus-based mutagenesis for the improvement of the stability of SaSrtAΔN59 WT: we generated multiple sequence alignments (MSA) to assess the eff ect of consensus mutations on the stability and activity of SaSrtAΔN59.[25] We attempted to further improve the stability of both WT and PM enzymes, while at the same time improving the enzyme’s activity. We applied consensus design[26,27] to search for evolutionary information preserved in homologous sequences.[28] Therefore, we built SrtA-MSAs
MagdaBinnenwerk.indd 79
80
consisting of a panel of homologous SrtAs derived from organisms adapted to different temperatures, such as mesophiles and thermophiles. Based on our MSAs, we selected and replaced eight non-consensus amino acids of SaSrtAΔN59 WT with the corresponding consensus ones. As a result, we found some mutations which not only led to an improvement of the thermostability of the protein, but also to an improvement of the enzymatic activity.[26,29] Finally, we introduced the two best mutations into the known SaSrtAΔN59 PM, making it even more active than previously reported[9] and less dependent on Ca2+ ions.[8]
MATERIALS AND METHODS
Consensus design
Sequences to be used in the MSA encoding class A sortases were retrieved from the UniProt database.[30] All sequences were verified for their proper distribution of soluble and insoluble domains using the TMHMM Server v. 2.0.[31] This server uses a hidden Markov model for the prediction of integral membrane proteins. As a result, 134 sequences encoding SrtA from different Gram-positive organisms were selected, with 35 enzymes from extremophile organisms. All organisms from which SrtA sequences were selected and used in the MSA, are shown in Figure 2. Prior to generation of the MSA using the PROMALS3D tool[32]
, secondary structure elements were aligned and highly conserved residues in SrtA were identified using high resolution 3D structures. To this purpose, S. aureus SrtA (PDB 1T2P), Bacillus
anthracis SrtA (PDB 2KW8), Streptococcus agalactiae SrtA (PDB 3RCC), S. pneumoniae
SrtA (PDB 4O8L), S. mutans SrtA (PDB 4TQX), Listeria monocytogenes SrtA (PDB 5HU4),
Corynobacterium diptheriae SrtA (PDB 5K9A), and Actinomyces oris SrtA (PDB 5UTT)
were selected. Based on predictions obtained from the structural alignment of these eight enzymes, all 134 retrieved sequences were used for the construction of the MSA.
Cloning, production, and purification of SaSrtAΔN59 WT, SaSrtAΔN59 PM and
selected mutants
The DNA sequence encoding SaSrtAΔN59 WT[33] was cloned into the pET28a plasmid (Novagen, USA) between the BamHI & NdeI restriction sites; the previously described mutations of SaSrtAΔN59 PM (P94R / D160N / D165A / K190E / K196T) and Ca2+-independent SaSrtA
ΔN59 mutant (E105K / E108Q, indicated as WT-Ca) were introduced via Quikchange® site-directed mutagenesis (Agilent, La Jolla, CA, USA). [34] After confirmation of the sequence, all plasmids were used for transformation
MagdaBinnenwerk.indd 80
81
5
of competent E. coli BL21(DE3) cells (New England Biolabs, USA). The SaSrtAΔN59 mutations selected from the MSA analysis were created via QuikChange® site-directed mutagenesis using templates encoding truncated SaSrtAΔN59 WT or SaSrtAΔN59 PM. The primers were designed according to the QuikChange® instruction manual and their sequences can be found in the supporting information, Table S1. After induction with 1 mM isopropyl β-d-1-thiogalactopyranoside (IPTG), protein
production was continued for 4 h at 37°C. Cells were then lysed by sonication and the cleared lysate was purifi ed using Ni-NTA resin (Qiagen, Hilden, Germany) according to the manufacturer’s instructions. All proteins were purifi ed to 95% purity via size exclusion chromatography using a Superdex75 HiLoad 16/60 column (GE Healthcare, MA, USA) in Tris-HCl buff er (50 mM, pH 7.5) supplemented with 150 mM NaCl and 10% v/v glycerol. Protein concentrations were determined using a Pierce bicinchoninic acid protein assay kit (Thermo Scientifi c).
Thermal stability
Thermal stability of all proteins was determined using a Jasco J-815 far-UV CD Spectropolarimeter equipped with a temperature controller, a thermostatted cell holder and a thermostatic bath. The instrument was under constant nitrogen fl ush. Experiments were performed in quartz cuvettes with a 1 mm path length. The protein concentration of WT SaSrtAΔN59 and the mutants was set to 1 mg/mL in a 50 mM Tris-HCl buff er, pH 7.5, supplemented with 150 mM NaCl. The secondary structure of SaSrtAΔN59 was recorded at 25°C as a spectrum between 210 and 250 nm; each measurement was repeated three times and corrected for solvent contributions. The molar ellipticity (θ) was calculated as reported before.[35] The reversibility of the unfolding process was assessed after subjecting the proteins to a temperature increase from 20°C to 70°C with a ramp of 1°C/min and cooling. Subsequently, thermal unfolding was monitored at the minimum measured at 210 nm, which corresponds to secondary structure content. Once spectra from the initial and the new measurement at 210 nm matched, thermodynamics of the samples could be calculated. Since spectral changes measured for SaSrtAΔN59 showed sigmoidal curves with a single transition from a native to an unfolded state, we assumed a two-state model for protein denaturation. The equilibrium constant of unfolding (Keq) was calculated from the fraction of folded and unfolded protein.[36] The enthalpy ΔHu for a globular monomeric protein was calculated using the Van ’t Hoff plot, and the Gibbs free energy of unfolding ΔGu was estimated using the Gibbs-Helmholtz equation. For calculation of the change in heat capacity upon unfolding
MagdaBinnenwerk.indd 81
82
ΔCp, the values of the coefficients were obtained through the relationship between reported parametric equations[37] and the amino acid sequence of the protein.
Library screening and fluorescence-based activity assay
The enzymatic activity of purified SaSrtAΔN59 WT and mutants was measured using a fluorometric assay as described previously.[38,39] Briefly, the cleavage of the internally quenched substrate Abz-LPETG-Dap(Dnp) (Bachem AG, Switzerland) (20 µM) was monitored using a FLUOstar Omega Spectrometer (BMG LABTECH), measuring the change in fluorescence at λem=460 nm after excitation at λex=355 nm at 37°C or 45°C. The NH2-5Gly-OH (Bachem AG, Switzerland) nucleophile was added to the reaction to a final concentration of 2 mM. The purified enzymes were individually added to a final concentration of 1 µM. Fluorescence signals were plotted against time and the slope values were obtained from the linear phase of the reaction.
To screen the mutant libraries, the activity of individual SrtA mutants was assessed in cell lysate. Briefly, E. coli BL21(DE3) transformants were grown overnight at 37°C in a 96-deep well Masterblock® (Greiner Bio-One), after which the overnight cultures were diluted 1:100 in 1 mL 2 YT media supplemented with Kanamycin (30 µg/mL) and incubated for 45 min at 37°C in an orbital shaker. Protein production was induced by addition of IPTG to a final concentration of 0.5 mM and continued for 4 h. After centrifugation, pellets were resuspended in 50 µL of BugBuster® (Novagen) supplemented with 1 mM EDTA and incubated for 30 min at room temperature with orbital shaking. Finally, 950 µL activity assay buffer with 5 mM CaCl2 was added to each well and the Masterblock® was centrifuged for 30 min at a speed of 3,000 rpm. The activity of the SrtA variants was measured using a fluorescence-based activity assay in 96-well black polystyrene plates (Greiner). Each well was filled with 50 µL of the cell lysate, assay buffer and internally quenched Abz-LPETG-Dap(Dnp) substrate to a final concentration of 20 µM. The expression level of the proteins was monitored by SDS-PAGE, as shown in the supporting information, Figure S1B. For the measurement of activity in the absence of Ca2+ ions, 5 mM CaCl
2 was exchanged for 5 mM ethylene glycol-bis(β-aminoethyl ether)-N,N,N′,N′-tetraacetic acid (EGTA). For the measurement of activity at 60°C, the enzymes were pre-incubated in a water bath at 60°C up for 90 min.
MagdaBinnenwerk.indd 82
83
5
RESULTS
Thermostability of Sortase WT and Sortase PM
For characterization of the secondary structure content and thermostability of the purifi ed SaSrtAΔN59 WT and SaSrtAΔN59 PM proteins, far UV-circular dichroism spectra were used. It showed minima at 210 nm which are characteristic for the presence of β-sheets. Additionally, the reversibility of unfolding was monitored by gradually bringing samples to a temperature of 70°C with steps of 1°C/min, then cooling down to 20°C and repeating the process; as shown in Figure 1, the unfolding and refolding transitions were found to be superimposable, which indicates that the proteins exhibited reversible unfolding.
FIGURE 1. Reversibility of protein unfolding measured by CD. Individual graphs represent the unfolding of native (red line) and unfolding of refolded (blue line) WT (left) and PM (right).[40] The
inset fi gures show the CD spectra of the native (red) and refolded proteins (blue).
The SaSrtAΔN59 WT was therefore corroborated to be a thermodynamically stable protein, refolding after being subjected to 70°C and cooled down to 25°C. The same behavior was observed for the SaSrtAΔN59 PM.
Identifi cation and selection of mutagenesis sites
Protein consensus design was performed to identify and select those amino acid positions in the structure of the SaSrtA∆59 WT, which could aff ect the thermostability of the enzyme. In the UniProt database, 166 SrtA sequences from diff erent
MagdaBinnenwerk.indd 83
84
bacteria were found. All sequences were then analyzed using the TMHMM Server; only sequences with a proper distribution of cytoplasmic, transmembrane and extracellular domains were used for further studies. Consequently, a total of 134 sequences were selected for the MSA. The taxonomic classifi cation of all organisms that carry the srtA gene is shown in Figure 2.
FIGURE 2. Organisms from which SrtA sequences were selected and used in the MSA. A. Pie chart representing the 17 diff erent taxonomic families of the 134 bacterial SrtA sequences selected from the UniProt database and analyzed using the TMHMM Server. B. Pie chart representing the classifi cation of 35 extremophile SrtA sequences. C. Pie chart representing the classifi cation of 35 bacterial SrtA sequences according to the extremophile group they belong to.
The 134 selected sequences from diff erent organisms were then classifi ed according to the organism’s habitat: extremophilic or mesophilic.[41] As shown in Figure 2B, the extremophile organisms showed taxonomical diversity. Most of these organisms were halophiles, while some were both halophilic and thermophilic, and others solely thermophilic (Figure 2C). Based on these categories, we developed two strategies: strategy 1 covered all 134 sequences encoding SrtA genes, while strategy 2 only covered the 35 SrtA sequences from extremophiles.
We used PROMALS3D to build the MSAs.[42] This progressive tool can be used to accurately construct alignments using information from available 3D structures, database homologs and predicted secondary structures.[32] For each amino acid the conservation index, ranging from 0 to 9 with 9 indicating the highest degree of conservation, was calculated with the AL2CO program[43] linked to the PROMALS3D server. We used the conservation index to assess the quality of our alignment and to select positions for the mutagenesis. The conservation index was used together with the consensus analysis to select amino acid substitutions (Figure 3).
MagdaBinnenwerk.indd 84
85
5
FIGURE 3. Part of the consensus design analysis of strategy 2, showing the positions selected for the mutagenesis in the SaSrtAΔN59 WT shown in green, with orange indicating the residues in the SaSrtAΔN59 WT. The catalytic Cys is highlighted in red with Cys184 residue in the SaSrtAΔN59 WT
depicted in yellow. For the consensus analysis, bold and uppercase letters represent conserved amino acid residues; “l” aliphatic residues (I, V, L); “h” hydrophobic residues (W, F, Y, M, L, I, V, A, C, T, H); “s” small residues (A, G, C, S, V, N, D, T, P); “+” positively charged residues (K, R, H); “t” tiny residues (A, G, C, S).
As expected, the conservation index for the Cys residue, which is the catalytic center of all SrtA enzymes, was the highest. Interestingly, the second highest conservation index after analyzing the srtA sequences of extremophile organisms was given to the amino acid residue in position 193, where all analyzed sequences except the SaSrtAΔN59 WT contained Arg. Some positions selected in this study had a conservation index below 5, but here consensus outcome was used as a decisive variable (Table 1). After analyzing the MSAs from both strategies described above, we identifi ed 29 non-conserved positions in the input SaSrtAΔN59 WT sequence (PDB structure 1T2P). More precisely, analysis of the 134 sequences selected for strategy 1 resulted in the identifi cation of 13 positions with non-conserved residues
MagdaBinnenwerk.indd 85
86
in SaSrtAΔN59 WT (supporting information Table S2). The advantage of this strategy was the considerable number of input sequences and high taxonomic diversity of the analyzed sequences. Analysis of the sequences selected for strategy 2 led to the identification of 16 positions at which SaSrtAΔN59 WT preferred a different amino acid compared to the consensus (supporting information Table S3). Closer examination of the MSA from strategy 2 showed that at position 121 all analyzed extremophile sequences showed a very strong preference for amino acids with positively charged side chains (Figure 3). Therefore, we substituted the polar, uncharged Thr in SaSrtAΔN59 WT with Arg, His and Lys. Of particular interest was position 193, since all analyzed sequences had an Arg residue at this position, whereas in the SaSrtAΔN59 sequence a hydrophobic Val was incorporated. We thus decided to introduce the specific point mutations T121R, T121K, T121H and V193R via Quikchange® mutagenesis instead of constructing libraries.
Next, we compared the results from both strategies and selected the six non-consensus residues in SaSrtAΔN59 WT that emerged in the MSAs from both strategies for further studies (Table 1). Since positions 91, 113, 129 and 198 showed a high variation in (types of) amino acids, site-saturation libraries were constructed on these positions. At the other two positions, 144 and 155, a more specific preference towards aliphatic amino acids was observed, and therefore these residues were mutated into the three aliphatic amino acids Ile, Val and Leu.
TABLE 1. Non-consensus residues in SaSrtAΔN59 WT identified by both strategies. Consensus
amino acids at positions 144 and 155 showed more specific preferences for amino acid substitutions. The other residues listed in this table were modified into all 20 natural amino acids.
Position Type of residue in the consensus sequence (Strategy 1 / Strategy 2) Modifications
P91 Hydrophobic / Hydrophobic All possible amino acids
Q113 Hydrophobic / Small All possible amino acids
Q129 Hydrophobic / Hydrophobic All possible amino acids
F144 Aliphatic / Aliphatic I, V, L
M155 Aliphatic / Aliphatic I, V, L
K198 Hydrophobic / Tiny All possible amino acids
Hydrophobic residues are W, F, Y, M, L, I, V, A, C, T, H; small residues are A, G, C, S, V, N, D, T, P; aliphatic residues are I, V, L; tiny residues are A, G, C, S.
MagdaBinnenwerk.indd 86
87
5
Enzymatic activity of the mutants
After purifi cation, SaSrtAΔN59 WT and single mutants (Table 1) were fi rst examined for their enzymatic activity on the quenched substrate. The three mutants made at position T121 were all found to be inactive. On the other hand, mutants M155V and V193R were found to be slightly more active than the WT (Figure 4).
Next, the libraries at positions 91, 113, 129 and 198 were screened for their activity using cell lysate (supporting information Figure S1). As a result, fi ve potential hits, P91T, Q113G, Q129I, Q129F and K198A, were found, which were purifi ed and assessed for their activity. Unfortunately, the purifi ed mutants showed lower activities than the WT (Figure 4). Finally, of the six mutants at positions F144 and M155, only mutant M155V was found to have an improved activity in comparison to the WT (Figure 4).
FIGURE 4. Bar graph showing the residual activity of purifi ed wild type (WT) and mutants selected from the MSAs (n=3). The white bars represent variants emerging from both strategy 1 and 2. The grey bars represent variants emerging from strategy 2. Variants M155V and V193R, showing higher activity than the WT, were selected for further analysis.
Mutations M155V and V193R were then incorporated in the previously described SaSrtAΔN59 PM, and the activity of the resulting mutants PM155 and PM193 was assessed at 37°C and 45°C. As shown in Figure 5, the temperature change did not have a signifi cant infl uence on the initial rate of the reaction. Interestingly, variant
MagdaBinnenwerk.indd 87
88
PM155 performed around 50% better than SaSrtAΔN59 PM at either temperature while the activity of variant PM193 was comparable to that of SaSrtAΔN59 PM.
FIGURE 5. Bar graphs showing the residual activity the mutants compared to the wild type (WT) (n=3). A. The enzymatic activity of single mutants M155V and V193R compared to WT, measured at 37°C (grey bars) and 45°C (white bars). B. The enzymatic activity of the SaSrtAΔN59 PM and PM
with incorporated single mutations (PM155 and PM193) compared to WT and measured at 37°C (grey bars) and 45°C (white bars).
In order to assess the thermostability at a temperature higher than 45°C we incubated the mutants for 90 min at 60°C, followed by measurement of the enzymatic activity (Figure 6). Both single mutants were more resistant to the increased temperature than the WT. Additionally, mutation V193R introduced into the PM made the mutant more thermostable. Unfortunately, mutation M155V did not have a similar eff ect on thermostability as V193R when introduced into the PM. Since the catalytic function of SaSrtAΔN59 has been described to rely on Ca2+ ions,[8,44] we examined the activity of mutants M155V and V193R, both exhibiting higher activity than the WT, in the presence and absence of this divalent ion. As a control for the activity of SaSrtAΔN59 WT and mutants in the absence of Ca2+ we used a known Ca2+-independent SaSrtA
ΔN59 mutant (E105K / E108Q, indicated as WT-Ca).[44] As shown in Figure 7, in the absence of the allosteric activator, single mutant M155V exhibited approximately a three-fold increase in activity in comparison to the WT. Interestingly, mutation M155V conferred Ca2+-independence to the PM, which may allow the use of this enzyme for in vivo conjugation reactions.
MagdaBinnenwerk.indd 88
89
5
FIGURE 6. Bar graph representing the residual enzymatic activity of the mutants after 90 minincubation at 60°C (n=3). The activity of enzymes before the temperature treatment was set as 100% (t = 0 min).
FIGURE 7. Bar graph showing the enzymatic activity of selected mutants in the absence of the Ca2+ ions (presence of EGTA) in relation to the Ca2+-independent mutant WT-Ca (n=3).
MagdaBinnenwerk.indd 89
90
Characterization of the thermostability
Mutants M155V, V193R, and PM155 and PM193 exhibiting enzymatic activities higher or comparable to SaSrtAΔN59 WT were examined for their thermostability using CD. The reversibility of unfolding was monitored by gradually subjecting samples to a temperature of 70°C with steps of 1°C/min, subsequently cooling down to 20°C and repeating the process; as shown previously for WT and PM, the unfolding and refolding transitions were found to be superimposable, which indicates that the proteins exhibited reversible unfolding (Figure 8).
FIGURE 8. Reversibility of protein unfolding as measured by CD (n=3). Individual graphs represent the unfolding of native (red line) and unfolding of refolded (blue line) single mutants M155V and V193R, and PM155 and PM193.[40] The inset fi gures show the CD spectra of the native (red) and
refolded proteins (blue).
MagdaBinnenwerk.indd 90
91
5
The fraction folded protein was calculated from the CD measurement and the ellipticity at 210 nm. The reversibility of the temperature-induced unfolding allowed the estimation of thermodynamic parameters from the CD measurement using the Van ’t Hoff equation. Slopes from the Van ’t Hoff plots, which gave R-squared values higher than 0.9, were used for the calculation of the thermodynamic parameters given in Table 2.
TABLE 2. Thermodynamic parameters of SaSrtAΔN59 WT and mutants (n=3).
Protein Tm(°C) ΔHu (kJ/mol) (kJ/mol)ΔGu WT 59 ± 0.2 224.9 ± 1.5 4.0 ± 0.2 M155V 56 ± 1.3 259.5 ± 3.5 8.4 ± 0.4 V193R 57 ± 0.5 270.8 ± 3.1 9.7 ± 0.3 PM 57 ± 0.1 234.5 ± 0.4 6.9 ± 0.1 PM155 54 ± 0.9 181.7 ± 3.1 3.1 ± 0.2 PM193 55 ± 0.8 271.6 ± 2.7 11.1 ± 0.8
Although a slight decrease in Tm values of the mutants was observed, almost all examined variants showed higher thermodynamic stability with increased ΔHu and ΔGu compared to WT. The only combination of mutations that led to reduced enzyme stability was found in the PM155 mutant. In contrast, the single modifi cation of position M155V led to an increased thermostability of the SaSrtAΔN59 WT. Nevertheless, the most prominent modifi cation of the SaSrtAΔN59 WT which increased both ∆Hu and ∆Gu was found at position V193 as a single mutation and in combination with the PM.
DISCUSSION
An overall stabilizing eff ect of a protein results from the presence of stabilizing and neutral, but also destabilizing mutations, which counterbalance each other.[45] As previously reported, thermophilic proteins have a higher frequency of the residues R and Y in their structures[46] as well as amino acids G, K and I[47] or G, A and V.[48] On the other hand, amino acids N, Q, M and C are considered to bring thermolability and are not often found in thermophilic proteins.[46,49] Nevertheless, thermal stability of a protein is not only dependent on the occurrence of particular amino acids, but also on the location of the residue selected for substitution as well as its interactions with other residues.[46]
MagdaBinnenwerk.indd 91
92
To broaden its range of applications, a protein’s stability can be engineered, either via the introduction of appropriate mutations or chemical modification at the surface of the protein.[50] Currently, with the advancement of computational techniques and DNA sequencing technologies, sequence-based approaches have gained interest and have led to the introduction of a consensus concept to this field of science.[27,45,51] The utility of this approach has been demonstrated by improving the stabilities of different proteins such as endoglucanases,[52] fluorescent proteins[53] and a maltogenic amylase.[54]
Although consensus design has been successfully applied to many proteins to improve their stability,[50] some key factors need to be considered before using this methodology. One of the most important aspects is the generation of an MSA. It requires utilization of high-quality sequences with well-conserved regions, similar length and no repetitions, which are not always available for the protein of interest. [55] Additionally, application of sequences belonging to different taxonomic groups can decrease the bias towards specific taxa or species. Apart from the quality, the number of sequences can also influence the accuracy of MSA, although an optimal number is not established.[56,57] In other words, both the quality and the quantity of sequences used for the generation of the MSA need to be properly selected. Our approach to the improvement of SaSrtA’s thermostability covers more than a hundred homologous input sequences belonging to nine families of bacteria available. Before generating the MSA, all duplicates were removed, and sequences were evaluated for their length.
In order to get a complete picture of SaSrtA’s stability, we calculated the Tm value, which is most commonly used for the estimation of thermal resistance. In addition, we determined ΔGu values, which take the difference in free energy of the folded and unfolded species into consideration.[58,59] Here, we corroborate that the structure of SaSrtAΔN59 WT is thermodynamically stable and exhibits a reversible unfolding (Figure 1). This is a great evolutionary advantage, which helps prevent protein aggregation in unfavorable conditions.[58] We decided to engineer the SaSrtAΔN59 WT and the SaSrtAΔN59 PM to potentially improve their thermostability: detection of destabilizing mutations and their removal can lead to an increase in thermostability.[45] We identified two thermodynamically more stable mutations, M155V and V193R (Figure 4 and Table 2). The V193R mutation had in fact the biggest impact on the thermostability of SaSrtA, and was found via strategy 2, for which only extremophiles were used to build the MSA. In comparison to WT, both M155V and V193R showed a more than twofold improved ΔGu, which could reduce the
MagdaBinnenwerk.indd 92
93
5
loss of activity over time. Indeed, mutants M155V and V193R did show a higher residual activity after incubation at 60°C compared to WT and even PM (Figure 6). One would expect a higher enzymatic activity at higher temperature[60]; interestingly, the mutants (alone and in combination with the PM) did not show an increased initial reaction rate (Figure 5). However, not all enzymes exhibit perfect Michaelis-Menten kinetics.[61] Although ΔG
u improved, ΔHu and Tm did not undergo a similar improvement (Table 2); we speculate this combination could lead to an improved stability without a further improvement in the enzymatic activity.[61]
Ca2+ is known to enhance the activity of SaSrtA.[4,33,62] Binding of the ion is coordinated through E105, E108 and E171, residues that are remote from the active site. Ca2+ binding combined with substrate binding leads to a closed conformation, which involves a disorder-to-order transition of the β6/β7 loop[4,63] and extensive structural changes such as shortening of part of the loop (T156 to V161). Previously, Ca2+ -independence was achieved by introducing mutations E105K and E108Q.[44] These mutations were also introduced into the PM; though the mutants were active in the absence of Ca2+, the activity was generally less than in the presence of the allosteric activator.[8,64,65] Interestingly, we found that mutation M155V rendered the enzyme activity in the absence of Ca2+ (Figure 7), even though M155V is located at the C-terminal end of strand β6 lining the substrate groove and not involved with Ca2+ binding. Ca2+-independence was conferred to PM155 as well. We hypothesize that mutations lining the substrate groove, such as M155V but also D160A and D165A in the PM, may improve substrate binding by altering the conformation of the β6/β7 loop, thereby also changing the role for the allosteric activator.
In summary, we explored the thermodynamic stability of S. aureus SrtA, and potential improvement thereof by means of consensus design. Consensus design was successfully applied to identify specifi c residues in the SaSrtAΔN59 WT sequence that did not only have an impact on thermal stability, but also on enzymatic activity, and even on Ca2+-dependence. We therefore anticipate that variants such as M155V and PM155 could fi nd an application in conjugation reactions in low Ca2+ environments, and potentially be used in in vivo labeling experiments.
ACKNOWLEDGMENTS
We would like to thank Dr. T.E. Adams (CSIRO Manufacturing, Parkville, Australia) for kindly providing the srtA gene. We acknowledge the support of dr. S. Romero-Romero with the thermodynamics experiments. The authors would also like to
MagdaBinnenwerk.indd 93
94
thank dr. C. Mayer for his help with CD experiments. This work was supported by a Human Frontiers Science Program (HFSP) long-term fellowship (LT001131/2011) and a Rosalind Franklin Fellowship (University of Groningen) to Y.L.B., and an Orange Tulip Scholarship provided to S.V.T.
MagdaBinnenwerk.indd 94
95
5
REFERENCES
1. S.K. Mazmanian, G. Liu, H. Ton-That, O. Schneewind, Science 1999, 285, 760–763. 2. A.W. Jacobitz, M.D. Kattke, J. Wereszczynski, R.T. Clubb, Adv. Prot. Chem. Struc. Biol.2017,
155, 223–264.
3. L.A. Marraffi ni, A.C. DeDent, O. Schneewind, Microbiol. Mol. Biol. Rev. 2006, 70, 192–221. 4. N. Suree, C.K. Liew, V.A. Villareal, W. Thieu, E.A. Fadeev, J.J. Clemens, M.E. Jung, R.T. Clubb,
J. Biol. Chem. 2009, 284, 24465–24477. 5. T. Proft, Biotechnol. Lett. 2009, 32, 1–10.
6. E.K. Call, T.R. Klaenhammer, Front. Microbiol. 2013, 4, 1–10.
7. Q. Wu, H.L. Ploegh, M.C. Truttmann, ACS Chem. Biol. 2017, 12, 664–673.
8. H.J. Jeong, G.C. Abhiraman, C.M. Story, J.R. Ingram, S.K. Dougan, PLoS One2017, 12, 1–15. 9. I. Chen, B.M. Dorr, D.R. Liu, Proc. Natl. Acad. Sci. 2011, 108, 11399–11404.
10. L. Chen, J. Cohen, X. Song, A. Zhao, Z. Ye, C.J. Feulner, P. Doonan, W. Somers, L. Lin, P.R. Chen, Sci. Rep.2016, 6, 1–12.
11. R. Chen, Trends Biotechnol. 2001, 19, 13–14.
12. F.H. Arnold, A.A. Volkov, Curr. Opin. Chem. Biol. 1999, 3, 54–59. 13. R.E. Cobb, R. Chao, H. Zhao, AIChE J. 2013, 59, 1432–1440.
14. K. Steiner, H. Schwab, Comput. Struct. Biotechnol. J. 2012, 2, e201209010. 15. S. Lutz, Curr. Opin. in Biotechnol.2011, 21, 734–743.
16. C. An, E.L. Chaikof, H.O. Ham, D.R. Liu, B.M. Dorr, Proc. Natl. Acad. Sci. 2014, 111, 13343– 13348.
17. J.M. Antos, M.C. Truttmann, H.L. Ploegh, Curr. Opin. Struct. Biol. 2016, 38, 111–118. 18. K. Steiner, K, H. Schwab, Comput. Struct. Biotechnol. J. 2012, 2, e201209010.
19. B.L. Xu, B.-L, M. Dai, Y. Chen, D. Meng, Y. Wangm N. Fang, X.F. Tang, B. Tang, Appl. Environ. Microbiol. 2015, 81, 6302–6313.
20. M.L. Bentley, E.C. Lamb, D.G. McCaff erty, J. Biol. Chem. 2008, 283, 14762 – 14771. 21. B.A. Frankel, Y. Tong, M.L. Bentley, M.C. Fitzgerald, D.G. McCaff erty, Biochemistry2007,
7269 – 7278.
22. Z. Li, J.D. Hirst, Chem. Sci.2017, 8, 4318–4333. 23. T. Tripathi, J. of Proteins and Proteomics2013, 4, 85–91.
24. M. Pelay-Gimeno, T. Bange, S. Hennig, T.N. Grossmann, Angew. Chem. 2018, 57, 11164-11170.
25. Z. Wu, H. Hong, X. Zhao, X. Wang, Bioresour. Bioprocess. 2017, 4. 26. B.T. Porebski, A.M. Buckle, Protein Eng. Des. Sel.2016, 29, 245–251. 27. M. Lehmann, M. Wyss, Curr. Opin. Biotechnol.2001, 12, 371–375.
28. C. Jäckel, J.D. Bloom, P. Kast, F.H. Arnold, D. Hilvert, J. Mol. Biol.2010, 399, 747–762. 29. B. Steipe, B. Schiller, A. Pluckthun, S. Steinbacher, J. Mol. Biol.1994, 240, 188-192. 30. UniProt Consortium, Nucleic Acids Res.2009, 37, D169–D174.
31. A. Krogh, B. Larsson, G. Von Heijne, E.L.L. Sonnhammer, E. L. L. J. Mol. Biol. 2001, 305, 567–580.
32. J. Pei, B.H. Kim, N.V. Grishin, Nucleic Acids Res. 2008, 36, 2295–2300.
33. U. Ilangovan, H. Ton-That, J. Iwahara, O. Schneewind, R.T. Clubb, Proc. Natl. Acad. Sci. 2001, 98, 6056–6061.
34. H. Liu, J.H. Naismith, J. H. BMC Biotechnol.2008, 8, 91. 35. N.J. Greenfi eld, Nat. Protoc. 2006, 1, 2876–2890. 36. I.M. Klotz, Proc. Natl. Acad. Sci. 1996, 93, 14411–14415.
37. D. Milardi, C. La Rosa, S. Fasone, D. Grasso, Biophys. Chem. 1997, 69, 43–51.
38. H. Ton-That, S.K. Mazmanian, K.F. Faull, O. Schneewind, J. Biol. Chem. 2000, 275, 9876– 9881.
39. M. Wójcik, N. Eleftheriadis, M.R.H. Zwinderman, A.S.S. Dömling, F.J. Dekker, Y.L. Boersma, Eur. J. Med. Chem. 2019, 161, 93–100.
MagdaBinnenwerk.indd 95
96
40. S. Romero-Romero, M. Costas, A. Rodríguez-Romero, D. Alejandro Fernández-Velasco, Phys. Chem. Chem. Phys. 2015, 17, 20699–20714.
41. P.H. Rampelotto, Life 2013, 3, 482–485.
42. J. Pei, M. Tang, N.V. Grishin, Nucleic Acids Res. 2008, 36, 30–34. 43. J. Pei, N.V. Grishin, Bioinformatics 2001, 17, 700–712.
44. H. Hirakawa, S. Ishikawa, T. Nagamune, Biotechnol. Bioeng. 2012, 109, 2955–2961. 45. M. Lehmann, L. Pasamontes, S.F. Lassen, M. Wyss, Biochim. Biophys. Acta - Protein Struct.
Mol. Enzymol. 2000, 1543, 408–415.
46. S. Kumar, C.J. Tsai, R. Nussinov, Protein Eng. Des. Sel. 2002, 13, 179–191. 47. S.T. Farias, M.C.M. Bonato, Genet. Mol. Res. 2003, 2, 383–393.
48. A.S. Panja, B. Bandopadhyay, S. Maiti, PLoS One 2015, 10, 1–21. 49. T.J. Taylor, I.I. Vaisman, BMC Struct. Biol. 2010, 10, S5.
50. H.P. Modarres, M.R. Mofrad, A. Sanati-Nezhad, RSC Adv. 2016, 6, 115252–115270. 51. M.W. Pantoliano, M. Whitlow, J.F. Wood, S.W. Dodd, K.D. Hardman, M.L. Rollence, P.N.
Bryan, Biochemistry 1989, 28, 7205–7213.
52. M. Anbar, O. Gul, R. Lamed, U.O. Sezerman, E.A. Bayer, Appl. Environ. Microbiol. 2012, 78, 3458–3464.
53. M. Dai, H.E. Fisher, J. Temirov, C. Kiss, M.E. Phipps, P. Pavlik, J.H. Werner, A.R. Bradbury Protein Eng. Des. Sel. 2007, 20, 69–79.
54. Y.W. Kim, J.H. Choi, J.W. Kim, C. Park, J.W. Kim, H. Cha, S.B. Lee, B.H. Oh, T.W. Moon, K.H. Park, Appl. Environ. Microbiol. 2003, 69, 4866–4874.
55. T.J. Magliery, J.J. Lavinder, B.J. Sullivan, Curr. Opin. Chem. Biol. 2011, 15, 443–451. 56. S.A. Jacobs, M.D. Diem, J. Luo, A. Teplyakov, G. Obmolova, T. Malia, G.L. Gilliland, K.T.
O’Neil, Protein Eng. Des. Sel. 2012, 25, 107–117.
57. B.T. Porebski, A.A. Nickson, D.E. Hoke, M.R. Hunter, L. Zhu, S. McGowan, G.I. Webb, A.M. Buckle, Protein Eng. Des. Sel. 2015, 28, 67–78.
58. D. Sanfelice, P.A. Temussi, Biophys. Chem. 2016, 208, 4–8.
59. T.A. Wright, J.M. Stewart, R.C. Page, D. Konkolewicz, J. Phys. Chem. Lett. 2017, 8, 553–558. 60. V.L. Arcus, E.J. Prentice, J.K. Hobbs, A.J. Mulholland, M.W. Van der Kamp, C.R. Pudney, E.J.
Parker, L.A. Schipper, Biochemistry. 2016, 55, 1681 – 1688. 61. R.M. Daniel, M.J. Danson, FEBS Lett. 2013, 2738 – 2743.
62. M.T. Naik, N. Suree, U. Ilangovan, C.K. Liew, W. Thieu, D.O.Campbell, J.J. Clemens, M.E. Jung, R.T. Clubb, J. Biol. Chem. 2006, 1817 – 1826.
63. X. Wang, J.L. Chen, G. Otting, X.C. Su, Sci Rep. 2018, 2 – 10.
64. M.D. Witte, T. Wu, C.P. Guimaraes, C.S. Theile, A.E. Blom, J.R. Ingram, Z. Li, L. Kundrat, S.D. Goldberg, H.L. Ploegh, Nat. Protoc. 2015, 508 – 516.
65. I. Wuethrich, J.G.C. Peeters, A.E.M. Blom, C.S. Theile, Z. Li, E. Spooner, H.L. Ploegh, C.P. Guimaraes, PloS One, 2014, 9.
MagdaBinnenwerk.indd 96
97
5
Supporting information for Sortase mutants with improved protein thermo-stability and enzymatic activity obtained by consensus design.
Magdalena Wójcik et al.
TABLE S1. Primers used for the preparation of mutants identifi ed using strategy 1 and 2.
Primer name Sequence (5’-3’)
SaP91X_F1 ccggtttatccgggtndtgcaacaccggaacag SaP91X_F2 ccggtttatccgggtvmagcaacaccggaacag SaP91X_F3 ccggtttatccgggtatggcaacaccggaacag SaP91X_F4 ccggtttatccgggttgggcaacaccggaacag SaP91X_R1 ctgttccggtgttgcahnacccggataaaccgg SaP91X_R2 ctgttccggtgttgctkbacccggataaaccgg SaP91X_R3 ctgttccggtgttgccatacccggataaaccgg SaP91X_R4 ctgttccggtgttgcccaacccggataaaccgg SaQ113X_F1 gaaagcctggatgatndtaatattagcattgccgg SaQ113X_F2 gaaagcctggatgatvmaaatattagcattgccgg SaQ113X_R1 ccggcaatgctaatattahnatcatccaggctttc SaQ113X_R2 ccggcaatgctaatatttkbatcatccaggctttc SaQ129X_F1 gatcgtccgaattatndttttaccaatctgaaagcagcc SaQ129X_F2 gatcgtccgaattatvmatttaccaatctgaaagcagcc SaQ129X_F3 gatcgtccgaattatatgtttaccaatctgaaagcagcc SaQ129X_F4 gatcgtccgaattattggtttaccaatctgaaagcagcc SaQ129X_R1 ggctgctttcagattggtaaaahnataattcggacgatc SaQ129X_R2 ggctgctttcagattggtaaatkbataattcggacgatc SaQ129X_R3 ggctgctttcagattggtaaacatataattcggacgatc SaQ129X_R4 ggctgctttcagattggtaaaccaataattcggacgatc SaF144X_For ggtagcatggtgtatnttaaagtgggtaatgaaacccgc SaF144X_Rev gcgggtttcattacccactttaanatacaccatgctacc SaM155X_F1 acccgcaagtataaandtaccagcattcgtgatg SaM155X_R1 catcacgaatgctggtahntttatacttgcgggt SaK198X_F1 gtgtgggaaaaacgcndtatttttgtggccacc SaK198X_F2 gtgtgggaaaaacgcvmaatttttgtggccacc SaK198X_F3 gtgtgggaaaaacgcatgatttttgtggccacc SaK198X_F4 gtgtgggaaaaacgctggatttttgtggccacc SaK198X_R1 ggtggccacaaaaatahngcgtttttcccacac SaK198X_R2 ggtggccacaaaaattkbgcgtttttcccacac SaK198X_R3 ggtggccacaaaaatcatgcgtttttcccacac SaK198X_R4 ggtggccacaaaaatccagcgtttttcccacac MagdaBinnenwerk.indd 97 MagdaBinnenwerk.indd 97 19/02/2020 13:54:0519/02/2020 13:54:05
98
Primer name Sequence (5’-3’)
SaT121R_For gaatattagcattgccggtcatcgctttattgatcgtccgaattatc SaT121R_Rev gataattcggacgatcaataaagcgatgaccggcaatgctaatattc SaT121K_For gaatattagcattgccggtcataaatttattgatcgtccgaattatc SaT121K_Rev gataattcggacgatcaataaatttatgaccggcaatgctaatattc SaT121H_For gaatattagcattgccggtcatcactttattgatcgtccgaattatc SaT121H_Rev gataattcggacgatcaataaagtgatgaccggcaatgctaatattc SaV193R_For gattataacgaaaaaaccggccggtgggaaaaacgcaaaatttttg SaV193R_Rev caaaaattttgcgtttttcccaccggccggttttttcgttataatc
FIGURE S1. A representative example of the results of the library screening and the expression level of the proteins A. Bar graph representing part of the screening results of the cell lysate assay of libraries of mutants in positions 91, 113, 129 and 198. Potential hits with higher activity than the WT were further analyzed. B. SDS-PAGE gel showing the expression level of the proteins from the cell lysate assay. The lanes of the gel correspond to the samples in the assay.
MagdaBinnenwerk.indd 98
99
5
TABLE S2. Results of the consensus analysis for strategy 1, considering all SrtA sequences.Position Residue in SaSrtA sequence Type of residue in the consensus sequence (134 sequences)
85 E Hydrophobic 91 P Hydrophobic 113 Q Hydrophobic 129 Q Hydrophobic 135 A Polar 144 F Aliphatic 151 R Hydrophobic 155 M Aliphatic 167 G Polar 194 W Small 198 K Hydrophobic 200 R Aliphatic 201 V Polar
Hydrophobic residues are W, F, Y, M, L, I, V, A, C, T, H; small residues are A, G, C, S, V, N, D, T, P; aliphatic residues are I, V, L; tiny residues are A, G, C, S.
MagdaBinnenwerk.indd 99
100
TABLE S3. Results of the consensus analysis for strategy 2, considering extremophile SrtA sequences. The two residues 121 and 193, selected for further studies, are depicted in red.
Position Residue in Sa-SrtA sequence Type of residue in the consensus sequence (extremophiles)
81 A Aliphatic 91 P Hydrophobic 113 Q Small 121 T Positively charged 129 Q Hydrophobic 136 A Aliphatic 144 F Aliphatic 155 M Aliphatic 160 D Hydrophobic 190 K Small 193 V R 195 E Aliphatic 196 K Aliphatic 198 K Tiny 199 I Polar 202 A Polar
Amino acid code: aliphatic (I,V,L), hydrophobic (W, F, Y, M, L, I, V, A, C, T, H), small (A, G, C, S, V, N, D, T, P), positively charged (K, R, H), tiny (A, G, C, S), polar (D, E, H, K, N, Q, R, S, T).
MagdaBinnenwerk.indd 100
